arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2006.09858cs.LGeess.SPstat.ML
收藏

几何相似性比较

Geometry of Similarity Comparisons

University of Basel(巴塞尔大学)

Puoya Tabaghi, Jianhao Peng, Olgica Milenkovic, Ivan Dokmanić

摘要:

许多数据分析问题可以转换为\emph{space forms} --欧氏空间、球面空间或双曲空间中的距离几何问题。通常,绝对距离测量通常是不可靠的或根本不可用的,并且只有以相似性的形式的绝对距离的代理是可用的。因此,我们提出以下要求:如果一组实体之间只有\emph{comparisons}的相似性,那么关于底层空间形式的几何学可以说些什么呢?为了研究这个问题,我们引入了目标空间形式的\textit{ordinal capacity}和相似性度量的\emph{ordinal spread}的概念.后者是测量中的复杂模式的指示器,而前者量化空间形式的容量以容纳具有特定有序分布配置文件的一组测量。证明了空间形式的序容量与其维数和曲率的符号有关。这导致了我们所称的相似图的欧几里得和球面嵌入维数的下界。更重要的是,我们表明,定义在相似图上的有序分布随机变量的统计行为可以用来识别其潜在的空间形式。我们支持我们的理论主张与实验加权树,单细胞RNA表达数据和球形制图测量。

英文摘要:

Many data analysis problems can be cast as distance geometry problems in \emph{space forms} -- Euclidean, spherical, or hyperbolic spaces. Often, absolute distance measurements are often unreliable or simply unavailable and only proxies to absolute distances in the form of similarities are available. Hence we ask the following: Given only \emph{comparisons} of similarities amongst a set of entities, what can be said about the geometry of the underlying space form? To study this question, we introduce the notions of the \textit{ordinal capacity} of a target space form and \emph{ordinal spread} of the similarity measurements. The latter is an indicator of complex patterns in the measurements, while the former quantifies the capacity of a space form to accommodate a set of measurements with a specific ordinal spread profile. We prove that the ordinal capacity of a space form is related to its dimension and the sign of its curvature. This leads to a lower bound on the Euclidean and spherical embedding dimension of what we term similarity graphs. More importantly, we show that the statistical behavior of the ordinal spread random variables defined on a similarity graph can be used to identify its underlying space form. We support our theoretical claims with experiments on weighted trees, single-cell RNA expression data and spherical cartographic measurements.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑